One of the more head-scratching themes of the Toronto Maple Leafs’ 2013-14 season was their penalty kill dropping from second best in the league the season prior to third from the bottom.

It was a big fall from grace for a team that returned its top four penalty killers (Jay McClement, Carl Gunnarsson, Dion Phaneuf, and Nikolai Kulemin), had excellent goaltending, and the same coaching staff. The team experienced a huge nose dive in almost every statistical category.

It is astonishing to look at:

Maple Leafs Penalty Kill Comparison - 2012-13 vs. 2013-14

SeasonPP GA/G4v5 GA PK Shots Against Times Shorthanded
2012-13.4 (2nd)18 (t-1st)157 (10th)157 (15th)
2013-14.72 (28th)49 (t-28th)438 (25th)268 (13th)

Other than the obvious curiosity to find out why the Leafs PK sunk like a stone, there are a few key factors that really pushed me to pursue this analysis. The first is this goal Blake Wheeler scored against the Leafs on the power play. I pointed it out at the time in a Leafs Notebook, but it was the first time it really caught my eye how deep the Leafs defencemen were along the blue line in their neutral zone system for the penalty kill. I didn’t elaborate at the time, but I did point out the obvious: [pull_quote_center]If a top power play unit can set up with ease, it’s going to score, period. This the NHL in the salary cap era – every team can throw out five players who are capable of passing, shooting, and scoring.[/pull_quote_center]

And this has stuck with me with ever since out of pure curiosity. It is tough to look these things up during the course of a season because of time constraints and other commitments, but in the summer, when things calm down, it is the perfect time to dig deeper.

The second factor was actually a quote from Jack Capuano of the Islanders, who said he was going to work on structural adjustments to his team’s 29th-ranked penalty kill:
[pull_quote_center]”We’re going to change the way we forecheck on it, and, when the puck is in our zone, we’re going to pressure differently.”[/pull_quote_center]
Without having the time to look at every single penalty kill the Leafs played through in the last two years to collect my data, I had to think of a logical way to use the resources I do have at hand to figure out how to go about analyzing the drastic shift in penalty killing proficiency that took place in a relatively short period of time. My first instinct was to collect as much video as I can on Leafs penalty kills in 2013 compared to the 2013-2014 season and record their process. Some things I looked for included: Who was generally starting the initial DZ faceoff, when were lines changed, was there a difference in formation between 2013 and 2013-2014, an emphasis on handedness, etc.

My second instinct was to look at every power play at 5v4 goal scored against the Leafs in the last two years and break them down to see if there was a variation in numbers or if the percentages held up in terms of types of goals allowed over the two years. Finding the video is easy and it gives me a basis of what has changed. I didn’t include 5v3 goals or 4v3 because it skews the data and the majority of power play goals occur at 5v4. I also think there is nothing really scientific or noteworthy to killing 5v3, as I view it to be a failing of the team on the power play more than anything (or as meaning the goalie stood on his head), although I’m open to arguments suggesting otherwise.

To this point, that is what I have done here in order to dig deeper as to understand the failings of the penalty kill. In turn, I think there is a reverse effect to be shown as well that can display what makes powerplays more effective. Because there is no baseline of information available, this can’t be taken as dogma, but it is a starting point onto which I will build as I search through other teams’ special team failings and successes in order to hopefully add to this and provide some context on what exactly has happened the last two years.

Breaking down powerplay goals against by category

I broke down power play goals into four categories to simplify it:

  1. The first category is a goal ten seconds off a zone entry, which is pretty self-explanatory.
  2. The second category is goals off a faceoff loss. I originally attempted “ten seconds” after a loss, but there were a few goals that were scored after the ten second mark and the Leafs had yet to touch the puck, so to me it logically made sense to expand that to count as long as possession was maintained and never switched teams.
  3. The third category is point shots, including tip-in goals but not rebounds. My reasoning is that a rebound goal presents an opportunity for a defensive player to make a play on the puck or clear a body and therefore it is not one in the same as a point shot (there were a decent amount of rebound goals that occurred on the second or third rebound as opposed to the first, too).
  4. The final category is goals ‘below the top of the circle,’ which does include the aforementioned rebounds, shots from the half wall, and passing plays. All of these goals are caused by defensive breakdowns that are often similar (i.e. someone lost a man/isn’t closing a lane down low), so I didn’t see the value in splitting them up and creating even more categories; if the value can be proven I’ll amend it, though.

NOTE: Goals off of faceoffs/right after zone entries could be off of point shots, passing plays, etc. but how they were originated is the focus here. In simple pie charts, here is what I have recorded so far: Toronto_Maple_Leafs_2012-2013-PPGAToronto_Maple_Leafs_2013-2014-PPGA So, what exactly has happened here? You’ll notice in both seasons, despite the Leafs percentages being polar opposite, they allowed a sizable amount of goals below the circle (set ups, rebounds, etc.). I don’t believe they got better at defending that last year; I think they just allowed more goals in other ways and it balanced it out because teams were setting up less. As I alluded to and suspected earlier, there were a lot more goals generated off of quick zone entries with a 10% hike. We often think of power plays traditionally with a team gaining the zone, setting it up, and working it around until they create a good scoring opportunity, but we see that a lot of goals are actually off of the rush or from getting a quick shot (which is funny because we usually consider players taking a quick shot off the rush on a PP “selfish”).

Shorthanded Faceoffs

What really interested me was the faceoffs. That was something that did not feel like an issue throughout the year, in the sense that it didn’t seem to get a lot of play in the media. What happened here? From my recordings, there was a big bulk of goals off of faceoffs between game 15 against Vancouver and game 31 against Boston. I recorded 6 goals off of faceoffs in that time, and for five of them one of Trevor Smith or Jerred Smithson was on.

Tyler Dellow did a lot of work on faceoffs at his now-defunct website, and in one piece he wrote:
[pull_quote_center]If you win an offensive zone faceoff, your Corsi% over the next 37 seconds will be about 74.5%. If you lose that faceoff, you’ll do the next 21 seconds at about 54.3%. So there’s a sizeable gap in there. That’s the value that winning the faceoff gives you.[/pull_quote_center]

Logically, that can only be magnified off a power play where you have a man advantage. McClement won the third most draws shorthanded in the league last year, but that number is a little deceiving as he won only 46.7% of the draws he took shorthanded. Bozak, by comparison, won 52.9% in the lockout year. This season, Bozak won 38.5% of his draws shorthanded, so their top two faceoff men shorthanded got manhandled. In the lockout year, McClement took the second most PK draws and went 50% (in fact, no other Leaf hit double digit shorthanded draws, so the Leafs on the whole were pretty darn good on the dot in the lockout).

A typical Leafs PK in the lockout year went something like this: Generally, Bozak would go out for the draw with one of McClement or Kulemin. After the draw Bozak went as hard as he could to get it out. Either he’d win it and they’d clear, or he’d pressure immediately knowing that he was getting off as soon as the puck went out. When Bozak was getting off, Kulemin or whoever was on would race down the ice and apply some pressure, plus a fresh penalty killer would be coming on and roughly 20-30 seconds would already be killed. If McClement was off, he’d hop on and stay for the majority of what was left of the kill (or the whole thing), and if Kulemin was on he’d switch with Komarov eventually, with Bozak coming on if there was another defensive zone faceoff. With that setup, Bozak was able to be very aggressive knowing that he was going off immediately.

Here are two goals he’s scored in the last two seasons; both were off of draws he lost, but he was aggressive and instantly challenged puck handlers:
This is a roundabout way of saying that it is not cut and dry that you need to win the draw or you are doomed. It is about what you do after you lose, too. You can lose the draw and retreat, or lose the draw, go through your man, and apply pressure. You can’t win every faceoff shorthanded, but Bozak knowing that he is going straight off allows him to go full throttle right off the hop. With McClement taking the draws and still knowing he was killing the remainder of the penalty regardless of if the puck was cleared or not, that was not the case. He was — maybe ‘conserving himself’ isn’t the best way to put it — but he essentially was saving energy (or simply didn’t have as much) to a certain extent. Carlyle in January even noted it when he said, “We think we’re overtaxing Jay McClement, playing him way too much. It shows in our penalty killing.”

Part 1 In Conclusion

We have spoken so much about the Leafs lack of depth throughout the summer, but one area we have not analyzed in depth is its effects on the penalty kill. Last season, the Leafs had JVR and Mason Raymond rounding out their top four penalty killers, despite them both playing heavy 5v5 minutes. It helped in the lockout that Kulemin and McClement played every game, while Bozak and Komarov missed a combined 8 games. Players like Hamilton, JVR and Steckel were sprinkled into the PK, but by and large everyone knew who the Leafs’ four guys were. In the 2013-2014 season, the Leaf had McClement and Kulemin again, but after that they were using glue and tape. Smithson got a shot, so did D’Amigo; eventually, they decided to just run JVR and Raymond while Bolland was hurt, but Bolland was also on the PK when he was healthy while Bozak saw his role reduced because he was getting crushed at the dot. We put a lot of stock into what PK system a team is running, but at the end of the day there is no magic bullet system to kill a penalty.

Every hockey player growing up learns how to play in a diamond and box, and teams run some sort of variation within them. The Leafs penalty kill has been bad pretty well every year since the 2005 lockout, and many times we have heard player X has been acquired and he is a good penalty killer (Jason Blake, Colby Armstrong, Kris Versteeg, Hal Gill, Pavel Kubina, etc). Did all these guys suddenly forget how to run a box-diamond PK? The more I dig into this, the more I am starting to think the key is to not have to do it as much at all. Where teams do differentiate themselves, as we can see in the data, is in the neutral zone to deny clean entry scoring opportunities off the hop, and in the faceoff circle, where teams can run you over if you are

A.) Losing an exorbitant amount of draws, and
B.) If you don’t pressure after the fact.

Faceoffs are, of course, not the only reason the Leafs penalty kill dropped dramatically, but it’s a good place to start.